Databases metabolite identification

Metabolism Metabolite identification Standard methods and databases Kerns et al., 1997... 

To use metabolic footprinting as a technique for high-throughput applications, benchmark spectra databases with identified peaks are required so that peak patterns obtained from MS or NMR analysis can be rapidly translated into relevant biological information. Common experimental procedures should, ideally, also be established for metabolite analysis [80] such as those existing in proteomics or transcriptomics. Nevertheless, the scientific community has only recently attempted to achieve these tasks. Several databases for identification of metabolomics signals by MS are now available, for instance, BIGG [81], BioCyc [82], MSlib [83], NIST [84], Metlin [85], and HMDB [86] databases. For a more comprehensive list of resources we refer to the review of Werner and coworkers [68]. 

Cui Q, Lewis lA, Hegeman AD, Anderson ME, Li J, Schulte CF, Westler WM, Eghbalnia HR, Sussman MR, Markley JL. Metabolite identification via the Madison Metabolomics Consortium Database. Nat. Biotechnol. 2008 26 162-164. 

Efforts focused on the automated analysis of 2D NMR spectra of complex metabolite mixtures are still somewhat limited. Considering the importance of peak alignment in 2D spectra, a statistical method was developed to align NMR peaks from the same metabolites across multiple 2D NMR spectra (120). A semiautomated method for metabolite identification from 2D TOCSY and HSQC spectra was also developed (103). The software tool, MetaboMiner, makes use of a spectral reference library extracted from publicly available databases to automatically match peaks and identify compounds. 

Cui Q, et al. Metabolite identification via the Madison Metabolomics Consortium Database. Nature Biotechnol 2008 26 162-164. 

C. Hildebrandt, S. Wolf, and S. Neumann. Database supported candidate search for metabolite identification. / Integr. Bioinf, 8(2) 157, 2011. 

In order to enable holistic metabolic assessment of living organisms, one requires methods that can acquire metabolic profiles in a rapid, reproducible and comprehensive manner, without bias towards compound classes. NMR meets these requirements effectively, and can assess metabolites down to the tens of pM level . This may seem high when compared to more targeted methods such as LC(MS) and GC(MS), but NMR has the imique advantage that hundreds of metabolites can be assessed in a reproducible and quantitative manner in a single shot. Metabolite identification is a notorious bottleneck in metabonomics. In anticipation of this challenge, we built a pH dependent AMIX H NMR spectral database for gut polyphenols fermentation products for which literature provided clues on their abundance in faeces and urine. 

KNApSAcK database for metabolite identification the signal intensities are sufficiently repeatable for input to statistical classification programmes. 

Computational methods including both metabolism databases and predictive metabolism software can be used to aid bioanalytical groups in suggesting all possible potential metabolite masses before identification by mass spectroscopy (MS) [116,117]. This approach can also combine specialized MS spectra feature prediction software that will use the outputs from databases and prediction software and make comparisons with the molecular masses observed... 

Figure 3.4 GC/MS metabolic profile of a polar M. truncatula root extract that provides the identification for many of the root components. Individual components are identified by matching their mass spectra to those in databases or by comparison with authentic samples. Using this approach we have identified a large number (>130 currently) of primary metabolites in M. truncatula.

Once a database is established, it is made available to other laboratories through the company s secured intranet, so that the information therein can be updated, retrieved and reviewed. The resulting structural library can be referenced throughout the lifetime of the drug for rapid identification of impurities, degradants, and metabolites. 

The 4-coumarate CoA ligase (4CL EC 6.2.1.12) enzyme activates 4-coumaric acid, caffeic acid, ferrulic acid, and (in some cases) sinapic acid by the formation of CoA esters that serve as branch-point metabolites between the phenylpropanoid pathway and the synthesis of secondary metabolites [46, 47]. The reaction has an absolute requirement for Mg " and ATP as cofactors. Multiple isozymes are present in all plants where it has been studied, some of which have variable substrate specificities consistent with a potential role in controlling accumulation of secondary metabolite end-products. Examination of a navel orange EST database (CitEST) for flavonoid biosynthetic genes resulted in the identification of 10 tentative consensus sequences that potentially represent a multi-enzyme family [29]. Eurther biochemical characterization will be necessary to establish whether these genes have 4CL activity and, if so, whether preferential substrate usage is observed. 

For this task, easily accessible properties of mixtures or pure metabolites are compared with literature data. This may be the biological activity spectrum against a variety of test organisms. Widely used also is the comparison of UV [90] or MS data and HPLC retention times with appropriate reference data collections, a method which needs only minimal amounts and affords reliable results. Finally, there are databases where substructures, NMR or UV data and a variety of other molecular descriptors can be searched using computers [91]. The most comprehensive data collection of natural compounds is the Dictionary of Natural Products (DNP) [92], which compiles metabolites from all natural sources, also from plants. More appropriate for dereplication of microbial products, however, is our own data collection (AntiBase [93]) that allows rapid identification using combined structural features and spectroscopic data, tools that are not available in the DNP. 

Use of an integrated system incorporating CCC separation, PDA detector, and LC-MS proved to be a valuable tool in the rapid identification of known compounds from microbial extracts.6 This collection of analytical data has enabled us to make exploratory use of advanced data analysis methods to enhance the identification process. For example, from the UV absorbance maxima and molecular weight for the active compound(s) present in a fraction, a list of potential structural matches from a natural products database (e.g., Berdy Bioactive Natural Products Database, Dictionary of Natural Products by Chapman and Hall, etc.) can be generated. Subsequently, the identity of metabolite(s) was ascertained by acquiring a proton nuclear magnetic resonance ( H-NMR) spectrum. 

The identification of the lipids is a very challenging task. The lack of comprehensive mass spectral libraries often limits the identification of compounds in LC-MS and shotgun methods. Some spectral libraries are available, such as the Human Metabolome Database (http /www.hmdb.ca), the METLIN Metabolite Database (http /metlin.scripps.edu) (24), and the MassBank (http /www. massbank.jp) (25). However, construction of universal spectral databases for API-MS is challenging due to the poor reproducibility and high interinstru-ment variability of fragmentation patterns. 

In the pharmaceutical industry, the techniques are being used to examine off-target effects particularly for the early identification of toxicity. MOA can be studied through metabolomics and can also be used as a quality control tool for complex mixtures such as foods or herbal medicines. Similarly, the tools and expertise of natural products chemists are essential in metabolomics, particularly in biomarker discovery (see also Volume 9). Biomarker discovery via untargeted metabolomics can lead to metabolite signatures (nuclear magnetic resonance (NMR) spectroscopy, mass spectrometry (MS), etc.) that are not present in current metabolomics databases. This is particularly true for plant secondary metabolism studies and nonmammalian metabolites. Structure elucidation then becomes critical to understanding the metabolomics results and for biomarker development. 

---